Mapping Jupiter from a Series of Photos

A few times I have sat outside and taken a series of pictures of Jupiter to create an animation of it rotating. Jupiter rotates so quickly that the motion is very apparent in these animations, and it’s a lot of fun to see in a sky that is usually very static. This got me thinking that during those evenings I see more than half of the planet, and I was curious if there was a way to display that.

This has been somewhat of an ongoing project for me (see the Planet Mapping section of the site) but it’s finally started to produce some results I’m proud of.

I decided to use this animation of Jupiter rotating from July 28th 2019. It’s only about an hour’s worth of rotation, but it’s the best quality I’ve managed for one of these videos.

Because the first and last image are only an hour apart I don’t expect to map a huge section of the planet, but this is a good proof of concept. During the summer and fall of 2020 I’ll have to take a longer video of Jupiter to make a better map.

These images were taken with my 10” dobsonian telescope, a 2x barlow lens, and a modified webcam. Due to an issue with the webcam firmware I had no way of actually knowing what the ISO or shutter speed was set at, I had to manually dial in the exposure.

My software takes an image of a planet, aligns it properly and then processes it into a Mercator projection. These are the first and last images of that video, and their resulting Mercator projections.

The next step is to combine the two maps into one larger map. There will be a large amount of overlap between the two, so I need to decide how I want to handle the overlapping pixels. Should I average them, take the brightest, take the first?

One issue that stands out is that the edges of the maps are very dark. If I were to average pixels you would still see this darkness. What I want is to take the brightest pixel when the two images overlap. This way hopefully the darker, less useful, pixels from the edge of the map will be replaced by brighter pixels from the center of the other map.

I started writing some software to handle this photo combination, until I realized this could be done in Photoshop or GIMP. Because I’m on a budget I don’t have Photoshop and figured out how to do it in GIMP. The trick is to load each image in separate layers, and use layer mode “brighten”. This mode will take the lightest of the overlapping pixels, exactly what I’m looking for. And the result was stunning:

I added some reference lines on the image to mark out 0,90,180,270 degrees longitude, as well as -45,0,45 degrees latitude. As you can see the image of Jupiter takes up more than half of the map. If I used only one image I would have exactly 180 degrees of longitude, but thanks to the second image taken an hour later I have more than 180 degrees of longitude filled in.

What’s even more fascinating is that now that we have a Mercator projection we can map it back to a three dimensional sphere in software.

My images may not be the most detailed, and there may be lots of similar maps already online, but I’m really proud to say that I was able to create a map of a planet from my own images.

From here I plan on taking more photos of Jupiter over a longer period of time to fill in the map, and try to create a map for Mars as well.